Advances in deep neural network approaches to speaker recognition

The recent application of deep neural networks (DNN) to speaker identification (SID) has resulted in significant improvements over current state-of-the-art on telephone speech. In this work, we report a similar achievement in DNN-based SID performance on microphone speech. We consider two approaches...

Descripción completa

Guardado en:
Detalles Bibliográficos
Publicado: 2015
Materias:
bottleneck features
channel mismatch
Deep neural networks
normalization
speaker recognition
Acceso en línea:https://bibliotecadigital.exactas.uba.ar/collection/paper/document/paper_15206149_v2015-August_n_p4814_McLaren
http://hdl.handle.net/20.500.12110/paper_15206149_v2015-August_n_p4814_McLaren
Aporte de:
Biblioteca Digital - Facultad de Ciencias Exactas y Naturales (UBA) de Universidad de Buenos Aires
id paper:paper_15206149_v2015-August_n_p4814_McLaren
record_format dspace
spelling paper:paper_15206149_v2015-August_n_p4814_McLaren2023-06-08T16:19:13Z Advances in deep neural network approaches to speaker recognition bottleneck features channel mismatch Deep neural networks normalization speaker recognition The recent application of deep neural networks (DNN) to speaker identification (SID) has resulted in significant improvements over current state-of-the-art on telephone speech. In this work, we report a similar achievement in DNN-based SID performance on microphone speech. We consider two approaches to DNN-based SID: one that uses the DNN to extract features, and another that uses the DNN during feature modeling. Modeling is conducted using the DNN/i-vector framework, in which the traditional universal background model is replaced with a DNN. The recently proposed use of bottleneck features extracted from a DNN is also evaluated. Systems are first compared with a conventional universal background model (UBM) Gaussian mixture model (GMM) i-vector system on the clean conditions of the NIST 2012 speaker recognition evaluation corpus, where a lack of robustness to microphone speech is found. Several methods of DNN feature processing are then applied to bring significantly greater robustness to microphone speech. To direct future research, the DNN-based systems are also evaluated in the context of audio degradations including noise and reverberation. © 2015 IEEE. 2015 https://bibliotecadigital.exactas.uba.ar/collection/paper/document/paper_15206149_v2015-August_n_p4814_McLaren http://hdl.handle.net/20.500.12110/paper_15206149_v2015-August_n_p4814_McLaren
institution Universidad de Buenos Aires
institution_str I-28
repository_str R-134
collection Biblioteca Digital - Facultad de Ciencias Exactas y Naturales (UBA)
topic bottleneck features
channel mismatch
Deep neural networks
normalization
speaker recognition
spellingShingle bottleneck features
channel mismatch
Deep neural networks
normalization
speaker recognition
Advances in deep neural network approaches to speaker recognition
topic_facet bottleneck features
channel mismatch
Deep neural networks
normalization
speaker recognition
description The recent application of deep neural networks (DNN) to speaker identification (SID) has resulted in significant improvements over current state-of-the-art on telephone speech. In this work, we report a similar achievement in DNN-based SID performance on microphone speech. We consider two approaches to DNN-based SID: one that uses the DNN to extract features, and another that uses the DNN during feature modeling. Modeling is conducted using the DNN/i-vector framework, in which the traditional universal background model is replaced with a DNN. The recently proposed use of bottleneck features extracted from a DNN is also evaluated. Systems are first compared with a conventional universal background model (UBM) Gaussian mixture model (GMM) i-vector system on the clean conditions of the NIST 2012 speaker recognition evaluation corpus, where a lack of robustness to microphone speech is found. Several methods of DNN feature processing are then applied to bring significantly greater robustness to microphone speech. To direct future research, the DNN-based systems are also evaluated in the context of audio degradations including noise and reverberation. © 2015 IEEE.
title Advances in deep neural network approaches to speaker recognition
title_short Advances in deep neural network approaches to speaker recognition
title_full Advances in deep neural network approaches to speaker recognition
title_fullStr Advances in deep neural network approaches to speaker recognition
title_full_unstemmed Advances in deep neural network approaches to speaker recognition
title_sort advances in deep neural network approaches to speaker recognition
publishDate 2015
url https://bibliotecadigital.exactas.uba.ar/collection/paper/document/paper_15206149_v2015-August_n_p4814_McLaren
http://hdl.handle.net/20.500.12110/paper_15206149_v2015-August_n_p4814_McLaren
_version_ 1768545481252667392

Ejemplares similares